chansung's picture
Model save
e773d8c verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9979123173277662,
"eval_steps": 500,
"global_step": 239,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0041753653444676405,
"grad_norm": 14820.435529068707,
"learning_rate": 1.2499999999999999e-05,
"loss": 60.135,
"step": 1
},
{
"epoch": 0.020876826722338204,
"grad_norm": 1628.1311380537702,
"learning_rate": 6.25e-05,
"loss": 50.6887,
"step": 5
},
{
"epoch": 0.04175365344467641,
"grad_norm": 3033.800304830903,
"learning_rate": 0.000125,
"loss": 42.1741,
"step": 10
},
{
"epoch": 0.06263048016701461,
"grad_norm": 6954.121232870651,
"learning_rate": 0.00018749999999999998,
"loss": 161.1634,
"step": 15
},
{
"epoch": 0.08350730688935282,
"grad_norm": 850.7915596439383,
"learning_rate": 0.00025,
"loss": 187.8921,
"step": 20
},
{
"epoch": 0.10438413361169102,
"grad_norm": 51.008981364388355,
"learning_rate": 0.0002999839868651235,
"loss": 27.0222,
"step": 25
},
{
"epoch": 0.12526096033402923,
"grad_norm": 41.28593832910529,
"learning_rate": 0.000299423886051382,
"loss": 40.1282,
"step": 30
},
{
"epoch": 0.14613778705636743,
"grad_norm": 94.54405893422229,
"learning_rate": 0.0002980665441538907,
"loss": 22.8182,
"step": 35
},
{
"epoch": 0.16701461377870563,
"grad_norm": 117.0436511760194,
"learning_rate": 0.0002959192031789579,
"loss": 19.1922,
"step": 40
},
{
"epoch": 0.18789144050104384,
"grad_norm": 23.24704464882293,
"learning_rate": 0.00029299332011978107,
"loss": 17.9784,
"step": 45
},
{
"epoch": 0.20876826722338204,
"grad_norm": 67.09115479950934,
"learning_rate": 0.0002893045058284311,
"loss": 13.938,
"step": 50
},
{
"epoch": 0.22964509394572025,
"grad_norm": 74.81050241825953,
"learning_rate": 0.00028487244172520246,
"loss": 12.678,
"step": 55
},
{
"epoch": 0.25052192066805845,
"grad_norm": 36.783670859376656,
"learning_rate": 0.0002797207747897198,
"loss": 11.5948,
"step": 60
},
{
"epoch": 0.27139874739039666,
"grad_norm": 42.67794856216472,
"learning_rate": 0.0002738769913940706,
"loss": 9.6311,
"step": 65
},
{
"epoch": 0.29227557411273486,
"grad_norm": 20.131225618062707,
"learning_rate": 0.0002673722706511174,
"loss": 8.6553,
"step": 70
},
{
"epoch": 0.31315240083507306,
"grad_norm": 16.501356621775457,
"learning_rate": 0.0002602413180604401,
"loss": 7.8996,
"step": 75
},
{
"epoch": 0.33402922755741127,
"grad_norm": 17.24824317039284,
"learning_rate": 0.00025252218033947993,
"loss": 7.1388,
"step": 80
},
{
"epoch": 0.35490605427974947,
"grad_norm": 23.28427284101539,
"learning_rate": 0.0002442560424278399,
"loss": 6.7571,
"step": 85
},
{
"epoch": 0.3757828810020877,
"grad_norm": 19.752003570940577,
"learning_rate": 0.00023548700774781242,
"loss": 6.4415,
"step": 90
},
{
"epoch": 0.3966597077244259,
"grad_norm": 16.62982088461208,
"learning_rate": 0.00022626186289353913,
"loss": 6.4479,
"step": 95
},
{
"epoch": 0.4175365344467641,
"grad_norm": 23.009578526601246,
"learning_rate": 0.0002166298280042877,
"loss": 6.1347,
"step": 100
},
{
"epoch": 0.4384133611691023,
"grad_norm": 9.902525054887667,
"learning_rate": 0.00020664229415371266,
"loss": 5.868,
"step": 105
},
{
"epoch": 0.4592901878914405,
"grad_norm": 18.872092670717567,
"learning_rate": 0.0001963525491562421,
"loss": 5.8684,
"step": 110
},
{
"epoch": 0.4801670146137787,
"grad_norm": 10.396862801643927,
"learning_rate": 0.00018581549325353126,
"loss": 5.6768,
"step": 115
},
{
"epoch": 0.5010438413361169,
"grad_norm": 6.643173772718449,
"learning_rate": 0.00017508734619791966,
"loss": 5.5378,
"step": 120
},
{
"epoch": 0.5219206680584552,
"grad_norm": 13.620597347405528,
"learning_rate": 0.00016422534729572738,
"loss": 5.4243,
"step": 125
},
{
"epoch": 0.5427974947807933,
"grad_norm": 12.675575738004211,
"learning_rate": 0.0001532874500107902,
"loss": 5.4237,
"step": 130
},
{
"epoch": 0.5636743215031316,
"grad_norm": 10.600479885243871,
"learning_rate": 0.00014233201275765494,
"loss": 5.3795,
"step": 135
},
{
"epoch": 0.5845511482254697,
"grad_norm": 14.207759763111122,
"learning_rate": 0.0001314174875341878,
"loss": 5.2682,
"step": 140
},
{
"epoch": 0.605427974947808,
"grad_norm": 10.28585644469174,
"learning_rate": 0.00012060210805487529,
"loss": 5.1446,
"step": 145
},
{
"epoch": 0.6263048016701461,
"grad_norm": 9.411071111902643,
"learning_rate": 0.00010994357904876106,
"loss": 5.0323,
"step": 150
},
{
"epoch": 0.6471816283924844,
"grad_norm": 4.20316220773891,
"learning_rate": 9.949876837974944e-05,
"loss": 4.9314,
"step": 155
},
{
"epoch": 0.6680584551148225,
"grad_norm": 6.642144688442386,
"learning_rate": 8.932340363194595e-05,
"loss": 4.9206,
"step": 160
},
{
"epoch": 0.6889352818371608,
"grad_norm": 7.2411357694832255,
"learning_rate": 7.947177477888472e-05,
"loss": 4.8218,
"step": 165
},
{
"epoch": 0.7098121085594989,
"grad_norm": 6.883273869392802,
"learning_rate": 6.999644452302975e-05,
"loss": 4.6982,
"step": 170
},
{
"epoch": 0.7306889352818372,
"grad_norm": 5.045481934015123,
"learning_rate": 6.0947967851014405e-05,
"loss": 4.6601,
"step": 175
},
{
"epoch": 0.7515657620041754,
"grad_norm": 4.832077424408084,
"learning_rate": 5.237462230091467e-05,
"loss": 4.5969,
"step": 180
},
{
"epoch": 0.7724425887265136,
"grad_norm": 4.500645176472541,
"learning_rate": 4.432215038069449e-05,
"loss": 4.548,
"step": 185
},
{
"epoch": 0.7933194154488518,
"grad_norm": 3.503698684667856,
"learning_rate": 3.6833515512134606e-05,
"loss": 4.5618,
"step": 190
},
{
"epoch": 0.81419624217119,
"grad_norm": 3.680197433632794,
"learning_rate": 2.9948672802388135e-05,
"loss": 4.5023,
"step": 195
},
{
"epoch": 0.8350730688935282,
"grad_norm": 2.8773090613106302,
"learning_rate": 2.3704355866196373e-05,
"loss": 4.5226,
"step": 200
},
{
"epoch": 0.8559498956158664,
"grad_norm": 2.5614101608626467,
"learning_rate": 1.813388083616068e-05,
"loss": 4.4463,
"step": 205
},
{
"epoch": 0.8768267223382046,
"grad_norm": 2.2102224877308676,
"learning_rate": 1.326696860675981e-05,
"loss": 4.3724,
"step": 210
},
{
"epoch": 0.8977035490605428,
"grad_norm": 2.056590025798714,
"learning_rate": 9.129586260518634e-06,
"loss": 4.3747,
"step": 215
},
{
"epoch": 0.918580375782881,
"grad_norm": 1.7949115931981623,
"learning_rate": 5.743808522387544e-06,
"loss": 4.3585,
"step": 220
},
{
"epoch": 0.9394572025052192,
"grad_norm": 1.5441796176488383,
"learning_rate": 3.1276999815337544e-06,
"loss": 4.3784,
"step": 225
},
{
"epoch": 0.9603340292275574,
"grad_norm": 1.549016859105234,
"learning_rate": 1.2952187089419642e-06,
"loss": 4.3766,
"step": 230
},
{
"epoch": 0.9812108559498957,
"grad_norm": 1.5003100935642444,
"learning_rate": 2.5614178506644934e-07,
"loss": 4.3115,
"step": 235
},
{
"epoch": 0.9979123173277662,
"eval_loss": 5.734466552734375,
"eval_runtime": 1.4349,
"eval_samples_per_second": 1.394,
"eval_steps_per_second": 0.697,
"step": 239
},
{
"epoch": 0.9979123173277662,
"step": 239,
"total_flos": 8193589641216.0,
"train_loss": 2.8353221366595025,
"train_runtime": 2240.0339,
"train_samples_per_second": 3.416,
"train_steps_per_second": 0.107
}
],
"logging_steps": 5,
"max_steps": 239,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 100,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 8193589641216.0,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}